AREA 1. MEMBRANE PROTEINS IN BIOENERGETICS The research focus in Area 1 has been the ATP synthase. This membrane-bound enzyme produces most of the ATP used by cells, and thus it is indispensable for most organisms. ATP synthases use a unique turbine-like rotary mechanism to facilitate the transmembrane flow protons down an electrochemical gradient, and harness the resulting energy gain to power the chemical reactions required for ATP production. Many key aspects of this fundamental process of energy conversion, however, remain poorly characterized - such as the molecular mechanism of the membrane domain of the enzyme. In FY15/16 we have continued our previous efforts to advance our understanding of this mechanism; specifically, we have focused on attaining novel structural information on the membrane domain of the enzyme by integrating a wide range of experimental data through advanced, rigorous computational methods - so as to establish a foundation for future mechanistic studies. In one study (see Leone et al., References) we utilized such methods to construct a model of the molecular structure of a subcomplex of the membrane domain of the enzyme - specifically of the two protein subunits that are known to be key for proton permeation and energy transduction, known as the c-ring and subunit-a. This model was developed to be not only objectively consistent with available electron-microscopy data, but also with an extensive sequence analysis of correlated mutations in both proteins, and with prior cross-linking and cysteine-accessibility experimental measurements. This systematic, integrative approach revealed unambiguously the topology of subunit-a, which had been previously unclear, as well as its relationship with the c-ring; indeed, our structure provides a clear rationale for a wealth of functional and biochemical data gathered for this enzyme over the last two decades, whose interpretation had been hindered by the lack of reliable structural information. Moreover, our structure reveals important clues in regard to the pathway followed by protons as they traverse the membrane through the complex, and explains the directionality of the rotary mechanism and its strict coupling to the electrochemical proton gradient. Lastly, the structure also provides insights into the mode of action of known inhibitors - e.g. antibiotics that selectively target the bacterial or mitochondrial enzymes. For example, mapping of mutations that confer resistance to oligomycin inhibition unexpectedly revealed that this antibiotic may bind to two distinct sites in the a-c interface, explaining its ability to block the mechanism of the enzyme irrespective of the direction of rotation of the c-ring. In summary, this study was a stepping-stone towards establishing the mechanism of the ATP synthase at the atomic level. In a second study (see Zhou et al., References), we also utilized computational methods to construct structural models of two prototypical c-rings at atomic resolution. Specifically, we modeled the c-rings of the ATP synthase from Saccharomyces cerevisiae and Escherichia coli, which are prototypical of the enzymes in mitochondria (e.g. human) and pathogenic bacteria, respectively. Our models capture the structures of these c-rings in a state in which they are loaded with protons, in the context of the membrane; that is, in the state that would mediate the recognition of a potential inhibitor or antibiotic. Thus, we anticipate that these structural models will be will be of value in future efforts to improve the potency and specificity of inhibitors of this class of essential enzymes. AREA 2. TRANSPORT AND SIGNALING ACROSS BIOLOGICAL MEMBRANES Two classes of membrane proteins of biomedical significance have been the focus in Area 2. First, the superfamily of cation/Ca2+ exchangers, which are key for the regulation of cellular calcium, e.g. during the heartbeat. Second, so-called TRPV channels, which are found in sensory neurons and play a central role, for example, in thermosensation. In the first project, we continued our investigations of the structure and mechanism of NCX_Mj, a prokaryotic homolog of the cardiac Na+/Ca2+ exchanger. Previously, we had succeeded in determining the structure of a Na+-bound state of this transporter. In FY15/16, we were able to also establish the structure of a Ca2+-bound state, in collaboration with Prof. Youxing Jiang at UT Southwestern/HHMI, who provided X-ray crystallographic data. With this information at hand, we were able to complete an ambitious, simulation-based study of the conformational mechanism of this exchanger. Like other membrane antiporters, NCX exchangers facilitate the uptake of one substrate, down its transmembrane electrochemical gradient, and utilizes this driving force to power the efflux of another, if necessary against its concentration gradient; in this case, the substrates are Na+ and Ca2+, respectively. This process, known as secondary-active transport, relies on a seemingly simple principle, known as the alternating-access hypothesis. This hypothesis postulates, first, that transporters such as NCX are able to cycle between conformations that expose the substrate binding sites to either side of the membrane in an alternative manner (i.e. not concurrently). And second, that the transition between these two open conformations does not occur unless the appropriate substrate occupancy states have been realized. Our theoretical study of NCX_Mj provided unprecedented insights into the nature of this mechanism. Specifically, our calculations demonstrated that binding of either Na+ or Ca2+ to the protein has the effect of progressively re-shaping its free-energy landscape, thus controlling its conformational dynamics. For example, we demonstrated that a mechanistically mandatory conformation that occludes the ion binding sites simultaneously to both sides of the membrane becomes energetically feasible only under two conditions, i.e. in two specific ion-occupancy states. The two specific states are precisely those with 3 Na+ or 1 Ca2+ bound, thus providing an elegant explanation for why this protein is an antiporter whose stochiometry is exactly 3:1 (see Liao, Marinelli et al., Bibliography). In the second project, we initiated a collaboration with Dr. Kenton Swartz at NINDS, in which aimed to gain structural insights into the mode by which TRPV1 channels become activated when exposed to toxins from venomous organisms, thus triggering pain signals. In this collaborative study, we determined the NMR structure of the DkTx tarantula toxin, and using available cryo-EM data and advanced molecular-modeling and computer-simulation methods, we modeled the atomic structure of the TRPV1 channel with and without the toxin bound. This computational work was combined and contrasted with electrophysiological and biophysical measurements from the Swartz lab. Our results revealed the mode in which DkTx binds to TRPV1 and its mechanical effect on the channel structure, leading to its activation. Interestingly, we found that the toxin interacts simultaneously with the channel and with the surrounding membrane, a mode of recognition not previously described. This study thus provided groundbreaking insights into how protein-protein interactions occur in a membrane environment, and into the temperature-dependent gating mechanism of a pharmacologically important membrane channel (see Bae, Anselmi et al., Bibliography). REFERENCES 1. Leone V, Faraldo-Gomez JD. Structure, mechanism, and inhibition of the membrane motor of the ATP synthase inferred from quantitative modeling. Under review. 2. Zhou W, Leone V, Faraldo-Gomez JD. Predicted structures of the proton-bound membrane-embedded rotor rings of the Saccharomyces cerevisiae and Escherichia coli ATP synthases. Under review.